AI Can't Recommend a Book It Can't Describe
The No Boring Books Method, updated for the machine that now reads your book first
What is the No Boring Books Method?
The No Boring Books Method is Julie Broad's five-layer framework for writing a non-fiction book that readers and AI answer engines can both describe and recommend. It builds a clear reader, brand, hook, out-of-book connection, and metadata — because a machine can only recommend a book it can explain.
Go ask ChatGPT for the best book on your topic. Use the words your ideal reader would actually type, not the words you'd use at a conference. Something like "what should I read if I just inherited a team that doesn't trust me?"
Now read the answer.
If your book isn't in it, you have a lot of company. Most non-fiction books aren't. And the reason is almost never the one authors assume.
Authors assume this is a technology problem. They think there's a setting they missed, a plugin they need, or some new kind of markup that tells the robots their book exists. So they go looking for the technical fix, and the technical fix doesn't move anything, and they decide AI is rigged against independent authors.
It isn't rigged. It's just doing what it does, which is answer a question using the books it can actually describe.
That's the whole thing. An AI can't recommend a book it can't describe. Ask it about leadership and it will name the books it can summarize in a sentence, place with a specific reader, and connect to a specific problem. A book with no clear reader, no clear claim, and a title that sounds like eleven other titles gives it nothing to work with. So it reaches past you and grabs something it can explain.
Here's the part that surprised me when I sat down to write this. I already wrote the fix. It's in Chapter 2 of Self-Promote & Succeed, published in 2023, and it's the same method both of my books are built on.
The method is five layers. Here they are, exactly as they appear in the book:
One: A connection to and understanding of your audience. Two: A clear brand that communicates your message to your ideal reader, so they care. Three: A juicy reason why that audience needs to read your book (the 'hook'). Four: Invitations that ask readers to form a relationship with you beyond the book. Five: SEO (search engine optimization) using metadata that positions your title, cover, and book copy for marketing success.
I used a cake to explain it, because a cake with a missing layer still comes out of the oven. It just isn't what anyone wanted. A book missing a layer still gets published. It just doesn't get found, and nobody can tell you why.
That's the No Boring Books Method in one line. Five layers, and a boring book is missing at least three of them.
When I wrote Layer Five, the search engine I had in mind was Amazon, with Google running second. That's changed. The search engine is now a language model, and readers are asking it in full sentences instead of typing three keywords. What hasn't changed is that a layer of your marketing was always built for machines. The machine got better at reading, so now it reads all five layers, not just the metadata one.
Same five layers. New reader. Let me take you through what each one means now.
01
Your Ideal Reader
How does an AI match your book to a real question?
If you try to write a book for everyone, it will be important to no one. Nobody asks an AI a general question. They ask, 'I'm a physician therapist starting a cash-pay practice, what should I read?' The specificity is in the question, so the specificity has to be in your book, your description, and your site. Write down the actual question your reader would ask a machine at 11 p.m. Then go look at your book description and see whether any of those words appear in it.
02
Your Author Brand
What makes you a citable source?
A model answering a question about your topic is going to reach for people it can identify. Who are you, what do you do, what have you done, and why should anyone listen. If that information lives only in your head and on the back of your book, it can't be picked up. If it's on your website, in your bio, in your byline on other people's sites, and in the way you're introduced on podcasts, it can. Consistency does more work here than volume. Being described the same way in twelve places beats being described twelve different ways.
03
The Hook
What sentence does the AI repeat?
The hook was always the most important thing. A hook is a claim. It says what this book argues and what the reader gets out of it, and it's the outcome of an outcome, not the surface benefit. When someone asks an AI why they should read your book, the hook is the answer it gives, assuming you gave it one. If your book has no claim, there's no question it answers, so there's nothing to retrieve it against. Can one sentence about your book make a stranger want it? If the sentence is 'it's a book about resilience,' you don't have a hook. You have a topic. Topics don't get recommended, claims do.
04
Your out-of-book connection
Where does third-party proof get made?
Layer Four is the invitation to keep the relationship going past the last page. Your newsletter, your reader magnet, your community, your talks. The AI angle is indirect and it matters more than the direct ones. Readers who stay connected are the readers who leave reviews, mention you in a forum, ask you onto a podcast, and argue about your idea somewhere public. All of that becomes text about your book written by someone who isn't you, which is the material a model trusts most. You don't get talked about by asking to be talked about. You get talked about by giving people a reason to stay in the room.
05
Metadata
Why is metadata still the hinge?
Everything I wrote about metadata still applies. Your title and subtitle, your categories, your BISAC codes, your keywords, your description, your reviews. The update is that a language model reads your description as prose, not as a keyword field. Stuffing it doesn't work anymore, and it never worked well. Writing a description that clearly states who the book is for, what problem it solves, and what the reader walks away with does work, for the machine and for the human, at the same time. That first sentence of your description is doing more work than any other sentence you own.
The 5 pillars
Where the 5 Pillars fit
If you've taken our free discoverability audit, you know the 5 Pillars of Discoverability: Positioning, Platform, Placement, Proof, and Public Discussion.
Here's how the two fit together, and it's simpler than it looks. The No Boring Books Method is what you build into the book. The five pillars are what you build around it. Layer Five is the hinge where the inside becomes findable on the outside. You can't fix a pillar problem with pillar tactics if the layer underneath it is missing. That's why authors run ads on a book with no hook and watch the money disappear. Fundamentals first, paid marketing second. That was true when the audience was human and it's true now.
Positioning
Platform
Placement
Proof
Public Discussion
The Name
Why the method is called No Boring Books
Most non-fiction books are boring because they lack a benefit-driven hook and any storytelling, even when the information inside them is good. I meant it as a reader problem. Boring books don't get finished, don't get recommended, and don't build anything for the author. It turns out to be the same problem from the machine's side. A boring book has no distinct reader, no claim, no specific stories, and no language of its own. There's nothing in it that a model can hold onto and repeat. Being generic used to cost you word of mouth. Now it costs you the recommendation too. The books that AI surfaces are specific, opinionated, and clearly about something for somebody. Which is the same list of things that make a book worth reading. That is why the method has the name it has, and why #NoBoringBooks was never only a slogan about writing style.
Your Move This Week
Pick one.
- Write out the exact question your ideal reader would ask an AI. Compare it to your book description, word for word.
- Say your hook out loud in one sentence. If it's a topic and not a claim, rewrite it.
- Google yourself and read your own bio the way a stranger would. Is it the same story everywhere?
The missing piece is always action. One of these, this week, beats a plan you never start.
If you want to know which layer is costing you the most, we do a free AI Discoverability Audit, a $200 value.